Summary: Retroviral vectors have been used in clinical gene therapy trials to effect permanent cure of inborn errors of metabolism, such as ADA-SCID, because these vectors can integrate into the host genome to provide permanent transgene expression in the targeted cells. Although the first generation of retroviral vector(LASN)has been useful in demonstrating the feasibility of gene therapy approaches, the relatively low transduction frequencies and immune responses against both selectable markers(neo gene)and the therapeutic protein itself, have limited its success. Thus, a second generation retroviral vector (MPSV-ADA) was constructed by removing the selectable marker (neo) and switching the MLV LTR with a MPSV LTR. As a result, selection of transfected packaging cell lines has become an extremely labor intensive process involving the screening of thousands of clones by RNA analyses. We first addressed the task of facilitating selection of transfected packaging cell lines by adopting the strategy of using a selectable marker, but placing it outside of the inter-LTR regions of the retroviral construct. Thus, the selection marker would be active in the packaging cells, but should not theoretically be incorporated into the retroviral transcript produced by the packaging cell lines. We created a novel viral construct (MPSV-GFP)in which a neo gene selection marker was directly cloned into the plasmid backbone of MPSV-GFP, yielding (MPSV-GFP)-neo. We have demonstrated that this is an effective strategy in that within three weeks of transfection, 45% of packaging cells transfected with (MPSV-GFP)-neo and raised on a high concentration of G418, expressed the construct, compared to only 8% of packaging cells that were co-transfected with separate MPSV-GFP and neo constructs. The viral titre from packaging cell lines expressing (MPSV-GFP)-neo were 2-3 fold higher than that of the co-transfected cells, in that 65-80% of HeLa cells were infected by supes from the (MPSV-GFP)-neo compared to 15-20% for (MPSV-GFP)+ (neo) expressing packaging cells(manuscript in preparation). We have further demonstrated both by DNA analysis and by functional studies, that the selection marker is not present in target cells transfected with retrovirus produced by the (MPSV-GFP)-neo packaging cell lines, indicating that tandem repeats were not incorporated in the retroviral transcript. The second project, in which we attempted to generate a retroviral construct that would be more highly expressed in lymphocytes has generated some interesting findings. We cloned an immunoglobulin kappa light chain enhancer into the inter-LTR region of (MPSV-GFP)and transfected both lymphocyte and non-lymphocyte cell lines. Incorporation of the B cell enhancer construct increased GFP expression in two human T cell lines (Jurkat and Molt-4) but, paradoxically, diminished transgenic expression in a human B cell line. Expression was also diminished in K562 cells, a pre-erythrocyte/macrophage cell line. The fidelity of expression in primary lymphocyte populations is under investigation.